Taming JSON with SQL: From Raw to Results
summary
Compares six systems—Hive, Drill, Spark, BigQuery, Redshift, Snowflake—by running the same JSON dataset through each and producing a three-row five-column example so readers see exact SQL patterns: Hive uses get_json_object and lateral view, Drill and Spark do schema inference and use flattened/explode, BigQuery requires a JSON schema file and auto-flattens one level, Redshift stores JSON as text and warns against large JSON in a column, and Snowflake shreds JSON into columns for near-relational performance while noting schema inference reads entire file and agility-versus-performance tradeoffs.
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker background
- 10:01Overview of JSON tools
- 10:45Working with JSON in Hive
- 13:41Querying JSON with Apache Drill
- 17:58Processing JSON in Spark
- 20:53BigQuery JSON handling
- 23:22JSON in Amazon Redshift
- 27:42Snowflake’s JSON variant type
- 30:09Key takeaways and performance tips
- 31:47Q&A session
related talks
01h 01mThe JSON evolution with SQL Database
51mDealing with JSON in the relational world
29mRetire your tables. It's all JSON now.
01h 06mSQL ❤️ JSON: The Best of Both Worlds
47mDealing with JSON in the Relational World
45mHow Well Do Relational Database Engines Support JSON?
55mThe JSON Evolution with SQL Database
46mThe JSON Evolution With SQL Database